On 27 Oct 2000, Dr. S. Shapiro wrote:
> I have quantified experimentally the activity ("X") of a
> half-dozen different products (A-F). These 6 commercial
> products all contain the same "active ingredient" over a
> range of different concentrations (and a couple of products
> share the same "active ingredient" concentration). The
> problem arises from the fact that these 6 commercial
> preparations each contain different additives that may
> possibly impact on the observed activity, and no two
> products contain identical sets of additives.
> Under these circumstances, is there a statistical procedure
> that wouldenable me to distinguish the effect of the
> "active ingredient" per se from the mitigating effects
> (either positive or negative) of the additives?
Under the conditions you describe, the short answer is "No".
> At first I thought of doing a simple oneway ANOVA of the 6
> products side-by-side, but then thought that this might
> not be able to distinguish between the effects of the
> "active ingredient" per se from that of the "active
> ingredient + additives".
That is correct. Unless you have _products_ that have the
active ingredient without additives, and other products that
have the active ingredient + additives, you have no way to
distinguish between these effects.
> Any further suggestions would be most welcome.
> I have access to two statistics software packages:
> MINITAB 11.21 (preferred) and Excel 97, but may be
> able to find other packages if I hunt around long enough.
Any statistics you need to do can be done in Minitab. But software
doesn't matter: the logical problem, at least as I understand you
to have described it, is intractable.
-- DFB.
----------------------------------------------------------------------
Donald F. Burrill [EMAIL PROTECTED]
348 Hyde Hall, Plymouth State College, [EMAIL PROTECTED]
MSC #29, Plymouth, NH 03264 (603) 535-2597
Department of Mathematics, Boston University [EMAIL PROTECTED]
111 Cummington Street, room 261, Boston, MA 02215 (617) 353-5288
184 Nashua Road, Bedford, NH 03110 (603) 471-7128
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
